Federated Learning Based on Stratified Sampling Optimization for Heterogeneous Clients
Federated learning(FL) is a new distributed learning framework for privacy protection, which is different from traditional distributed machine learning: 1)differences in communication, computing, and storage performance among devices(device heterogeneity),2)differences in data distribution and data...
Saved in:
Published in | Ji suan ji ke xue Vol. 49; no. 9; pp. 183 - 193 |
---|---|
Main Authors | , , , , |
Format | Journal Article |
Language | Chinese |
Published |
Chongqing
Guojia Kexue Jishu Bu
01.09.2022
Editorial office of Computer Science |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Federated learning(FL) is a new distributed learning framework for privacy protection, which is different from traditional distributed machine learning: 1)differences in communication, computing, and storage performance among devices(device heterogeneity),2)differences in data distribution and data volume(data heterogeneity),and 3)high communication consumption.Under heterogeneous conditions, the data distribution of clients varies greatly, which leads to the decrease of model convergence speed.Especially in the case of highly heterogeneous condition, the traditional FL algorithm cannot converge and the training loss curve will fluctuate greatly with the increase of local iterations.In this work, a FL algorithm based on stratified sampling optimization(FedSSO) is proposed.In FedSSO,a density-based clustering method is used to divide the overall client into different clusters.Then, some available clients are proportionally extracted from different clusters to participate in training.Therefore, various data are |
---|---|
AbstractList | Federated learning(FL) is a new distributed learning framework for privacy protection, which is different from traditional distributed machine learning: 1)differences in communication, computing, and storage performance among devices(device heterogeneity),2)differences in data distribution and data volume(data heterogeneity),and 3)high communication consumption.Under heterogeneous conditions, the data distribution of clients varies greatly, which leads to the decrease of model convergence speed.Especially in the case of highly heterogeneous condition, the traditional FL algorithm cannot converge and the training loss curve will fluctuate greatly with the increase of local iterations.In this work, a FL algorithm based on stratified sampling optimization(FedSSO) is proposed.In FedSSO,a density-based clustering method is used to divide the overall client into different clusters.Then, some available clients are proportionally extracted from different clusters to participate in training.Therefore, various data are Federated learning(FL) is a new distributed learning framework for privacy protection,which is different from traditional distributed machine learning:1)differences in communication,computing,and storage performance among devices(device heterogeneity),2)differences in data distribution and data volume(data heterogeneity),and 3)high communication consumption.Under heterogeneous conditions,the data distribution of clients varies greatly,which leads to the decrease of model convergence speed.Especially in the case of highly heterogeneous condition,the traditional FL algorithm cannot converge and the training loss curve will fluctuate greatly with the increase of local iterations.In this work,a FL algorithm based on stratified sampling optimization(FedSSO) is proposed.In FedSSO,a density-based clustering method is used to divide the overall client into different clusters.Then,some available clients are proportionally extracted from different clusters to participate in training.Therefore,various data are involved |
Author | Lu, Chen-yang Zhou, Hao-hao Deng, Su Wu, Ya-hui Ma, Wu-bin |
Author_xml | – sequence: 1 givenname: Chen-yang surname: Lu fullname: Lu, Chen-yang – sequence: 2 givenname: Su surname: Deng fullname: Deng, Su – sequence: 3 givenname: Wu-bin surname: Ma fullname: Ma, Wu-bin – sequence: 4 givenname: Ya-hui surname: Wu fullname: Wu, Ya-hui – sequence: 5 givenname: Hao-hao surname: Zhou fullname: Zhou, Hao-hao |
BookMark | eNotTj1PwzAU9FAkSunOGIk5xd8fI1SUVqrUoRVii5zYrhxSOzipBPx6DEVveO_u3p3uBkxCDBaAOwQXCEnFH9qhff9cYAwZhJiTCZiifJSIiLdrMB8GX0NMOM2DpuB1ZY1NerSm2Fqdgg_H4kkPGcZQ7MeseOcz2utT3_2Ku370J_-d-fzgYirWdrQpHm2w8TwUy87bMA634MrpbrDz_z0Dh9XzYbkut7uXzfJxWxpFZWldg43kDFOORdMoyLUyzFFGja5rQzBHXNa5J3UcOqExppQYRWoBjbRMkxnYXGJN1G3VJ3_S6auK2ld_REzHSqfRN52tsBKidphlH6OCGC04V5AohTXP2OWs-0tWn-LH2Q5j1cZzCrl9hQXCUkqmJPkBf_NsvA |
ContentType | Journal Article |
Copyright | Copyright Guojia Kexue Jishu Bu 2022 |
Copyright_xml | – notice: Copyright Guojia Kexue Jishu Bu 2022 |
DBID | 7SC 8FD JQ2 L7M L~C L~D DOA |
DOI | 10.11896/jsjkx.220500263 |
DatabaseName | Computer and Information Systems Abstracts Technology Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional DOAJ Directory of Open Access Journals |
DatabaseTitle | Computer and Information Systems Abstracts Technology Research Database Computer and Information Systems Abstracts – Academic Advanced Technologies Database with Aerospace ProQuest Computer Science Collection Computer and Information Systems Abstracts Professional |
DatabaseTitleList | Computer and Information Systems Abstracts |
Database_xml | – sequence: 1 dbid: DOA name: DOAJ Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Computer Science |
EndPage | 193 |
ExternalDocumentID | oai_doaj_org_article_2977bf258e55473da766903992a6473f |
GroupedDBID | -0Y 2B. 2C0 5XA 5XJ 7SC 8FD 92H 92I ACGFS ALMA_UNASSIGNED_HOLDINGS CCEZO CUBFJ CW9 GROUPED_DOAJ JQ2 L7M L~C L~D TCJ TGT U1G U5S |
ID | FETCH-LOGICAL-d948-efc2d86524627cc906a9d5f454dabbd326168b4614f60f7a22443d93b70d8e5a3 |
IEDL.DBID | DOA |
ISSN | 1002-137X |
IngestDate | Thu Jul 04 21:09:39 EDT 2024 Fri Sep 13 08:18:58 EDT 2024 |
IsDoiOpenAccess | true |
IsOpenAccess | true |
IsPeerReviewed | false |
IsScholarly | true |
Issue | 9 |
Language | Chinese |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-d948-efc2d86524627cc906a9d5f454dabbd326168b4614f60f7a22443d93b70d8e5a3 |
OpenAccessLink | https://doaj.org/article/2977bf258e55473da766903992a6473f |
PQID | 2712888598 |
PQPubID | 2048282 |
PageCount | 11 |
ParticipantIDs | doaj_primary_oai_doaj_org_article_2977bf258e55473da766903992a6473f proquest_journals_2712888598 |
PublicationCentury | 2000 |
PublicationDate | 2022-09-01 |
PublicationDateYYYYMMDD | 2022-09-01 |
PublicationDate_xml | – month: 09 year: 2022 text: 2022-09-01 day: 01 |
PublicationDecade | 2020 |
PublicationPlace | Chongqing |
PublicationPlace_xml | – name: Chongqing |
PublicationTitle | Ji suan ji ke xue |
PublicationYear | 2022 |
Publisher | Guojia Kexue Jishu Bu Editorial office of Computer Science |
Publisher_xml | – name: Guojia Kexue Jishu Bu – name: Editorial office of Computer Science |
SSID | ssib023646461 ssib017479302 ssib051375750 ssib001164759 ssj0057673 |
Score | 2.3305075 |
Snippet | Federated learning(FL) is a new distributed learning framework for privacy protection, which is different from traditional distributed machine learning:... Federated learning(FL) is a new distributed learning framework for privacy protection,which is different from traditional distributed machine... |
SourceID | doaj proquest |
SourceType | Open Website Aggregation Database |
StartPage | 183 |
SubjectTerms | Algorithms Clients Clustering Convergence Decay rate Federated learning federated learning|privacy protection|clustering|stratified sampling|distributed optimization|convergence analysis Heterogeneity Machine learning Optimization Sampling |
Title | Federated Learning Based on Stratified Sampling Optimization for Heterogeneous Clients |
URI | https://www.proquest.com/docview/2712888598/abstract/ https://doaj.org/article/2977bf258e55473da766903992a6473f |
Volume | 49 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV09T8MwELVQJxa-EYWCPLCGpv7OSCuqCgkYKKhbZMc2UKBBpJUQv55z4qqVGFiYosRDlLvz3bNy9x5C50YYpyD3J9YYnTCfponWhiROWlmkXkhb66fc3IrRA7ue8Mma1FfoCWvogRvDdQkAFOMJV44HmVyrpYADXaBT1QLufZ19e3ztMFUDgUCTtSrUALshDldEV4E1XawRqfEelYBb0mUOBxAum9b8IP1B5WT5g1Nlojutpq9fF2FANRxgaCT7_5XL6wI13EFbEVniy-aLdtHG9_Me2l6qNuC4iffR4zDwRwDEtDiSqz7hPtQyi8sZrslqXzzgUnyvQ7M5LN5BVnmP45oYMC4ehRaaEiLPlYsKD97CSGV1gMbDq_FglER1hcRmTCXOF8QqwQkTRBZFlgqdWe4ZZ1YbYwHV9YQyYCPmReqlhlLPqM2okakFX2h6iFqzcuaOEDap90QJG1QVWSaIlrZgisJFUVcI10b9YKH8o-HPyAOjdf0A_JxHP-d_-bmNOkv75nGbVTmRUF6V4pk6_o93nKBNEqYb6hayDmrNPxfuFDDH3JzV4fUDnS3Lgw |
link.rule.ids | 315,786,790,870,2115,27957,27958 |
linkProvider | Directory of Open Access Journals |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Federated+Learning+Based+on+Stratified+Sampling+Optimization+for+Heterogeneous+Clients&rft.jtitle=Ji+suan+ji+ke+xue&rft.au=LU+Chen-yang%2C+DENG+Su%2C+MA+Wu-bin%2C+WU+Ya-hui%2C+ZHOU+Hao-hao&rft.date=2022-09-01&rft.pub=Editorial+office+of+Computer+Science&rft.issn=1002-137X&rft.volume=49&rft.issue=9&rft.spage=183&rft.epage=193&rft_id=info:doi/10.11896%2Fjsjkx.220500263&rft.externalDBID=DOA&rft.externalDocID=oai_doaj_org_article_2977bf258e55473da766903992a6473f |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1002-137X&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1002-137X&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1002-137X&client=summon |